ARTICLE pubs.acs.org/est
Effect of a Nonionic Surfactant on Biodegradation of Slowly Desorbing PAHs in Contaminated Soils Marisa Bueno-Montes,† Dirk Springael,‡ and Jose-Julio Ortega-Calvo*,† † ‡
Instituto de Recursos Naturales y Agrobiología, C.S.I.C., Apartado 1052, E-41080-Seville, Spain Division Soil and Water Management, Catholic University of Leuven, Kasteelpark Arenberg 20, B-3001 Leuven, Belgium.
bS Supporting Information ABSTRACT: The influence of the nonionic surfactant Brij 35 on biodegradation of slowly desorbing polycyclic aromatic hydrocarbons (PAHs) was determined in contaminated soils. We employed a soil originated from a creosote-polluted site, and a manufactured gas plant soil that had been treated by bioremediation. The two soils differed in their total content in five indicator 3-, 4-, and 5-ring PAHs (2923 mg kg-1 and 183 mg kg-1 in the creosotepolluted and bioremediated soils, respectively) but had a similar content (140 mg kg-1 vs 156 mg kg-1) of slowly desorbing PAHs. The PAHs present in the bioremediated soil were highly recalcitrant. The surfactant at a concentration above its critical micelle concentration enhanced the biodegradation of slowly desorbing PAHs in suspensions of both soils, but it was especially efficient with bioremediated soil, causing a 62% loss of the total PAH content. An inhibition of biodegradation was observed with the high-molecular-weight PAHs pyrene and benzo[a]pyrene in the untreated soil, possibly due to competition effects with other solubilized PAHs present at relatively high concentrations. We suggest that nonionic surfactants may improve bioremediation performance with soils that have previously undergone extensive bioremediation to enrich for a slowly desorbing profile.
’ INTRODUCTION A major factor limiting the success of bioremediation of polycyclic aromatic hydrocarbons (PAHs) in polluted sites is the low bioavailability of these compounds, due to slow desorption from the solid matrix.1,2 The resulting residual concentrations are of strategic value because they can limit future uses of the sites after treatment. Innovative methods are needed, not only to predict the behavior of these recalcitrant pollutant fractions in soils, but also to increase their bioavailability for an enhanced bioremediation performance. Desorption of PAHs from contaminated soils is well represented by a biphasic pattern, where a first phase of fast desorption (rate constant >0.1 h-1) is followed by a much slower phase, with half-lives that may reach months or years.3 Slow desorption can be prominent if soils are enriched in compartments such as semisolid, NAPL-like creosote materials, which limit biodegradation due to a restricted surface area and a slow diffusion inside the organic phase,4,5 or black carbon-type materials, with a strong affinity for sorption of PAHs.6 The desorption pattern of PAHs can be seriously affected when the pollutants remain in soils for long time periods, due to aging phenomena that increase the pool of slowly desorbing chemicals, with the subsequent reduction in biodegradation rates.7 Based on an extensive literature review on desorption, solubilization, and/or biodegradation performance r 2011 American Chemical Society
of various surfactants,8-13 it is conceivable that nonionic surfactants can increase biodegradation rates of slowly desorbing PAHs in soils. In general, the nonionic surfactant group is the most frequently used in biodegradation studies, mainly due to the absence of electrical charge in the molecule of surfactant, what minimizes eventual toxic effects. Furthermore, in comparison with cationic or anionic surfactants, the nonionic group also shows in general a lower critical micelle concentration (CMC). Despite its potential in bioremediation, the capacity of nonionic surfactants to enhance biodegradation rates of slowly desorbing PAHs remains relatively unexplored. Whereas the promoting effect of nonionic surfactants on solubilization of PAHs from soils is a well-known phenomenon,12,13 studies reporting precise measurements of biodegradation rates for slowly desorbing PAHs in the presence of surfactants are very scarce. The only available assessment was recently provided with the nonionic surfactant Brij 30 for contaminated soil from a manufactured gas plant site (MGP) that had first been treated in an aerobic bioreactor.14 By using a single-step, slurry incubation during 18 Received: October 22, 2010 Accepted: February 21, 2011 Revised: February 14, 2011 Published: March 04, 2011 3019
dx.doi.org/10.1021/es1035706 | Environ. Sci. Technol. 2011, 45, 3019–3026
Environmental Science & Technology days, the study evidenced surfactant-enhanced desorption and biodegradation of residual PAHs in the bioremediated soil, whereas the surfactant had no effect on biodegradation in the untreated soil. However, the magnitude of the slowly desorbing fraction of PAHs present in the soils was not determined. We employed for this research two contaminated soils; one of them originated directly from a creosote-polluted site, whereas the other was a MGP soil that had been previously treated extensively by bioremediation. The soils differed in their total content in PAH but had a similar content of slowly desorbing PAH, as determined by Tenax extraction. According to the literature 8-13 and our own experience,15,16 we selected Brij 35, an nonionic alkylpoly (ethylene glycol) ether surfactant, for our study. We determined the effect of the surfactant at a concentration above its CMC on biodegradation of slowly desorbing fractions of PAHs in these soils through a dual radiorespirometry/residue analysis method, what allows precise estimations of compound bioaccessibility.16,17
’ MATERIALS AND METHODS Chemicals. [9-14C]-phenanthrene (13.1 mCi mmol-1),
[1,2,3,4,4a,9a-14C]-anthracene (20.6 mCi mmol-1), [3-14C]fluoranthene (45.0 mCi mmol-1) and [4,5,9,10-14C]-pyrene (58.7 mCi mmol-1), all of them with a radiochemical purity >98%, were provided by Sigma Chemical Co., St. Louis, MO. The nonionic alkyl poly(ethylene glycol) ether surfactant Brij 35 [C12H25O(CH2CH2O)23H] was purchased from Sigma-Aldrich (Steinheim, Germany). Tenax (60-80 mesh, 177-250 μm) was supplied by Chrompack. Soils. Two soils were used in this study: a creosote-polluted soil and a bioremediated soil. The former was a clay soil, classified as a calcaric fluvisol, provided by EMGRISA (Madrid, Spain) from an abandoned wood-treating facility in Andujar (Jaen, Spain), with a record of pollution by creosote going back 100 years. This soil was obtained from a point with heavy pollution (4501 mg kg-1 total PAH, sum of 16 EPA PAH), and had 3% organic matter and 60% clay. A detailed description of this soil is given elsewhere.15 The bioremediated soil originated from a Danish manufactured gas plant (MGP) site. It was obtained from a remediation company (Soilrem, Kalundborg, Denmark) that had treated the soil in biopiles during several years to bring the total PAH concentration down to 275 mg kg-1 (16 EPA PAH). These residual PAHs exhibited a high resistance to dissipation, according to further bioremediation efforts performed by the company that included organic amendments and composting.2 The soil had 3.6% organic matter, 28.8% clay and a pH of 7.96. These soils were air-dried, sieved (2 mm mesh) and kept in glass containers in the dark at 4 °C until use. Desorption. The procedure for desorption experiments is similar to the one previously described.3 Briefly, 0.5 g dry soil, 35 mL milli-Q water, 0.2 mL formaldehyde (40%), and 0.7 g Tenax TA beads were placed in 50 mL stainless steel centrifuge tubes equipped with a stainless steel sealing (Heraeus-Sorvall, Madrid) and kept at 23 ( 2 °C and 120 rpm on an rotary shaker. After certain time intervals, the tubes were centrifuged for 10 min at 17 212g. The floating Tenax beads were completely recovered with a spatula without disturbing the soil pellet. Then, the same amount of fresh Tenax beads was added to the tubes, the soil pellet was resuspended, and the cycle was repeated. The Tenax recovered from the tubes was extracted with 100 mL of acetone/hexane (1:1) in 250 mL screw-capped
ARTICLE
Erlenmeyer flasks that were kept for 5 h on a rotary shaker operating at 120 rpm. The extract was evaporated to near dryness, redissolved in acetonitrile, and filtered. PAH analysis was performed by HPLC as described below. The total mass of PAH desorbed plus the amount still present in the soil at the end of the experiments was for phenanthrene and pyrene higher than 98% in the creosote-polluted soil. Recoveries were in all other cases higher than 80%, with the exception of fluoranthene (69.3%) and benzo[a]pyrene (76%) in the creosote-polluted soil and anthracene (58.1%) in the bioremediated soil. Desorption data could be empirically described by the following first-order, two-compartment kinetic model:1 St =So ¼ F fast expð - kfast tÞ þ F slow expð - kslow tÞ
ð1Þ
where St and So (g) are the soil-sorbed amounts of PAHs at time t (h) and at the start of the experiment, respectively, Ffast and Fslow are the fast- and slow-desorbing fractions, and kfast and kslow (h-1) are the rate constants of fast and slow desorption. The values of Ffast, Fslow, kfast, and kslow were obtained by minimizing the cumulative squared residuals between experimental and calculated values of ln (St/So). The software used for the minimization was Microsoft Excell 2003 (Solver option). Surfactant Sorption onto Soil and Solubilization of Soil PAHs by Surfactant Micellar Solutions. Sorption of the surfactant onto the soil was assessed by measuring the increase in critical micelle concentration (CMC) in soil suspensions, as compared to aqueous solution. The CMC was calculated as the lowest surfactant concentration not leading to a significant decrease in surface tension of an inorganic salts solution (pH 5.7, described elsewhere17). The CMC in soil suspensions was determined with supernatants obtained after centrifuging (17 212g, 15 min) suspensions of creosote-polluted soil containing 1 g/70 mL inorganic salts solution, that were equilibrated with different surfactant concentrations during 24 h in a rotary shaker. Surface tension was determined at 25 °C with a TD1 Lauda ring tensiometer and plotted as dyn/cm against the logarithm of surfactant concentration in mg/L. The solubilization of PAHs in the presence of aqueous-phase Brij 35 was determined in 50 mL steel centrifuge tubes having a suspension that contained 0.5 g dry soil, 35 mL inorganic salts solution, 0.2 mL formaldehyde (40%), and 5 mL of a Brij 35 solution to give a final surfactant concentration of 20 g/L. The suspensions were maintained under the same conditions as Tenax desorption experiments. After certain time intervals, the tubes were centrifuged for 10 min at 17 212g, and an aliquot of the supernatant was analyzed for PAHs. The remaining supernatant was carefully decanted without disturbing the soil pellet, fresh surfactant solution was added to the tube, and the washing cycle was repeated. The complete procedure involved at least five washing steps performed during 168 h. Biodegradation. The microbial accessibility of native PAHs was assessed through a dual 14C/residue analysis method.17 For this aim, 1 g of dry soil was placed in 250 mL Erlenmeyer flasks, and mixed with 1 mL of a sterile, inorganic salts solution, that contained between 80 000 and 100 000 dpm of the corresponding 14C-labeled PAH (phenanthrene, anthracene, fluoranthene, and pyrene - benzo[a]pyrene was not expected to be mineralized under the experimental conditions used). According to the specific activity of the stocks, the label added corresponded to a negligible fraction of the native PAH-load in the soils. The 14Clabeled PAH were added completely dissolved in the aqueous phase, a procedure that guaranteed the homogenization of the 3020
dx.doi.org/10.1021/es1035706 |Environ. Sci. Technol. 2011, 45, 3019–3026
Environmental Science & Technology
ARTICLE
Table 1. Desorption Kinetics, As Obtained with Tenax, And Solubilization by an Anionic Surfactant (Brij 35, 20 g L-1) of Polycyclic Aromatic Hydrocarbons (PAHs) from Polluted Soils creosote-polluted soil (Andujar site) PAH concentration PAH
-1
(mg kg )
kfast
a
-1
(h )
kslow
a
-3
-1
(10
h )
Ffast
a
(%)
bioremediated soil from manufactured gas plant site (Soilrem) FBrij35b (%)
PAH concentration -1
(mg kg )
kfasta -1
(h )
kslowa -4
(10
-1
h )
Ffasta
FBrij35b
(%)
(%)
phenanthrene
1329.5 ( 52.7
0.38 ( 0.00 2.70 ( 0.03 96.1 ( 0.1 102.0 ( 2.4
42.3 ( 5.6
0.16 ( 0.00
1.6 ( 0.1
12.0 ( 1.6 62.3 ( 0.3
anthracene fluoranthene
286.3 ( 9.1 881.5 ( 3.4
0.26 ( 0.04 2.69 ( 1.70 94.6 ( 0.6 107.0 ( 1.2 0.30 ( 0.02 2.40 ( 0.58 95.6 ( 1.1 93.2 ( 2.6
15.4 ( 1.5 54.8 ( 4.3
0.13 ( 0.02 0.14 ( 0.00
1.2 ( 0.1 1.9 ( 0.1
9.6 ( 1.8 32.0 ( 1.4 18.3 ( 4.0 35.2 ( 10.6
pyrene
395.2 ( 15.0
0.26 ( 0.03 2.20 ( 0.50 92.3 ( 1.1
99.3 ( 1.0
48.3 ( 6.7
0.15 ( 0.01
1.5 ( 0.0
15.1 ( 1.9 32.2 ( 5.2
30.7 ( 4.4
0.11 ( 0.03 1.80 ( 0.20 86.7 ( 0.9
NDc
22.4 ( 0.1
0.06 ( 0.00
1.3 ( 0.1
13.7 ( 1.4 23.8 ( 2.8
benzo[a]pyrene
Kinetic parameters for desorption as obtained with Tenax extraction. b PAH fraction extracted with Brij 35 (20 g L-1) after 168 h. c ND, not determined. a
labeled compounds within the soil. Then, more inorganic salts solution was added to complete a final volume of 70 mL. The solution contained, when appropriate, Brij 35 to give a final concentration of 20 g L-1. The flasks were then closed with Teflon-lined stoppers, from which a 2 mL vial containing 1 mL of 0.5 M NaOH was suspended to trap 14CO2. The flasks were incubated at 23 ( 2 °C on an orbital shaker operating at 100 rpm. Mineralization of 14C-PAH in these biostimulated soil suspensions was then measured as described elsewhere.17 Mineralization rates were calculated and compared as the slope of the regression lines drawn with the points belonging to the phase of maximum mineralization. The final extents of mineralization were not considered in our bioaccessibility assessment. Instead, residual contents of native PAH were determined in separate, duplicate flasks which were incubated under the same conditions but contained no 14C-labeled compound. At the end of the assay (mineralization rate
